Phase Space AI

03 Valuation Analysis

Cooper Companies [COO]

The Cooper Companies [COO] — Valuation, Street Bridge & Factor Scorecard

investment-memo v1.6.0 · analysis dated 2026-07-28, migrated onto the Criteria framework 2026-07-29

Spot: $72.085 (scan) / $72.06 (Alpaca close in the body) · Shares 195.03m · Archetype COMPOUNDER Implied path: 6.1% required vs 7.3% demonstrated → margin +1.3pp (PASS) · 12-month target $77 base (+7.1%)

This document no longer states a position verdict. Under v1.6.0 the memo scores every Criteria, blocks on none, and emits an analysis. Whether that analysis justifies a position is a question about a particular book.

---

0. Valuation under v1.6.0 — two horizons, both mandatory

Inputs are taken from reports/scan/COO_analysis.json (as_of 2026-07-28) rather than recomputed, per the update spec. Where the scan contradicts a figure below, the scan wins and the correction is stated.

Note the enterprise-value discrepancy, corrected rather than overwritten silently. The body of this document uses EV $16,375m (net debt $2,321.4m at 30-Apr-26). The scan computes EV $15,779m from spot $72.085 × 195.03m shares less net debt $1,720.7m. The $600m difference is the net-debt definition (the scan nets more cash/short-term investments). §0.1 uses the scan's $15,779m, which is what the pre-computed required CAGR was solved against.

0.1 The implied-path test (reverse DCF) — the Valuation Criteria · PASS, margin +1.3pp

Sensitivity — over the exit multiple, never over scenario probabilities:

Breakeven is a 21.4x exit multiple against 22.7x today — the price survives a further 6% multiple compression before the demonstrated growth rate stops covering it.

DISCLOSED DEFECT IN THE PRE-COMPUTED INPUT, AND IT FLIPS THE RESULT (2026-07-29). The update spec describes a "derived terminal margin" in the scan file. There is none — COO_analysis.json has no terminal_margin key. The published required_cagr_pct inverts exactly to a flat 20.0% terminal EBIT margin applied to every name in the universe, the hardcoded 0.20 that coverage_scan.py's own comments record as a defect fixed after these files were written.

COO's own operating margin is 16.7%, well below that 20.0% — and this name is the one where it matters. Substituting 16.7%: required CAGR 10.0%, margin −2.6pp, breakeven exit multiple 25.6x against 22.7x today. The Criteria flips from PASS to FAIL.

Terminal EBIT margin	Source	Required CAGR	Margin	Result
20.0%	scan's universe-wide constant (undisclosed in the file)	6.1%	+1.3pp	PASS
16.7%	COO's own current operating margin	10.0%	−2.6pp	FAIL

The scan figure is reported as the headline because the spec instructs use of the pre-computed inputs. The correction is material and the reader should treat COO as sitting on the PASS/FAIL boundary, not comfortably inside PASS. The direction of the correction is unambiguous: a 16.7%-margin business valued at a 20.0% terminal margin is being flattered. This is the same failure class the scan's own comment describes — distributors on ~1% margins scoring negative required growth — in a milder form.

The −2.6pp corrected figure is also what reconciles COO to the batch expectation that these four names sit "roughly −2 to −9pp" on the new method.

0.2 The 12-month target · $77 base, +7.1% to spot

Built from near-term consensus and named events, with the multiple anchored on COO's own trading history and percentile — never a peer median projected years forward.

Step 1 — near-term revenue. Alpha Vantage EARNINGS_ESTIMATES, cached 2026-07-28, 15 analysts. COO's fiscal year ends 31 October, so trailing revenue at July-2027 is a quarter of FY2026 plus three quarters of FY2027:

Step 2 — named events inside the window, each dated in the Catalyst Criteria calendar: the CooperSurgical strategic-review update (management indicated it may land before the FQ3-26 print, early September 2026); FQ4-26 results and FY2027 guidance (~4 December 2026); MiSight category-share data. No undated catalyst is asserted here, and the estimated dates carry the (est.) flag they carry in the calendar.

Step 3 — the multiple, on COO's own history. Price-to-sales on a constant current share count (195.03m) against an internally-consistent EDGAR-derived TTM revenue series, daily, 2021-01-04 → 2026-07-28 (n = 1,397).

COO's multiple sits at the 5th percentile of five and a half years of its own history — and, unlike the longer windows, at the median of the last six months. Read together: a large de-rate that has stopped de-rating.

(The 6-month median equals today's multiple exactly, so the 75th percentile is used for the up-case rather than a degenerate one. The 3-year median of 5.12x would imply $119, +65%, and is not used: it belongs to a pre-de-rate regime and reverting to it inside twelve months is not a claim this document will make.)

The base case contains no multiple opinion — it is consensus revenue growth at an unchanged multiple.

Sanity band vs the external reference. The Street target dispersion cited in §4 is $61 – $103. The band above sits inside it.

CORRECTION — the old target was $77.45 (+7.5%), from probability-weighted scenarios. The new base is $77 (+7.1%). On this name the two instruments happen to agree almost exactly; the change is still real, because the sensitivity now runs over the multiple rather than over the scenario probabilities.

0.3 Data hygiene and corrections to this document

Nothing below this line has been deleted. §§1–6 are the original body. Where they state a conclusion this framework no longer draws, a superseding note marks it.

---

1. Where the model stands, and what it ties to

Full detail in 02_Financial_Model_Notes.md. Headline outputs:

Every actual-year line ties to the filed statement. The tie-out block is in the workbook and every check returns 0.0: revenue, gross profit, operating income and net income for FY2023A/FY2024A/FY2025A against the 10-K; total assets, total liabilities, current assets and current liabilities against the FY2025 balance sheet; cash from operations and capex for FY2024A and FY2025A against the filed cash-flow statement. The balance sheet also balances (residual 0.2, a disclosed rounding/restatement-vintage plug) — but the tie-out block, not the balance check, is what establishes that the inputs are right.

FY2026E revenue of $4,303.5m is the company guidance midpoint issued 4 June 2026 ($4.285–4.321bn); FY2026E non-GAAP EPS of $4.62 is the guidance midpoint ($4.58–4.66). H1 FY2026 is filed actual.

---

2. DCF

WACC build. Risk-free 4.69% (10Y UST proxy). ERP 5.00%. Raw 252-day beta vs SPY 0.558 (computed from Alpaca SIP daily log returns; beta vs XLV is 0.994 — COO is a market-insensitive, sector-sensitive stock). Blume-adjusted beta 0.33 + 0.67 × 0.558 = 0.704. Cost of equity 8.21%. Pre-tax cost of debt 4.02% (FY2025 interest expense $100.0m over average debt ≈ $2,490m); after-tax 3.18%. Equity weight 85.1%. WACC = 7.46%. Terminal growth 2.50%.

Cash tax rate is used for the DCF, not the book rate: FY2025 cash taxes paid $100.0m over pre-tax income $566.5m = 17.7%. The model ramps 19.0% → 21.5% over the forecast.

The FY2026 working-capital inflow and FY2027 outflow are the litigation accrual and its cash settlement; they net to approximately zero across the horizon.

Sensitivity (WACC × terminal growth):

(Grid values are computed independently in the chart-pack heatmap using the same UFCF stream; the $87.74 headline uses the workbook's exact mid-year discount factors, which is why the $84 grid cell differs slightly. The point is the shape, not the third significant figure.)

Honest reading of this DCF. 76.7% of the value is terminal, the implied exit multiple (13.0x) is above where the stock trades today (12.3x FY2026E), and the value swings from $59 to $110 across a defensible WACC/g box that straddles the spot price. A DCF on a 4%-growth company with a 7.5% WACC is an argument about the discount rate, not about the business. It is reported because the framework requires it; it is not what the decision rests on.

---

3. Comparables

On FY2026E: EV/revenue 3.8x, EV/adjusted EBITDA 12.3x, P/E on non-GAAP EPS 15.6x (FY2026E) and 14.5x (FY2027E), FCF yield 5.5%.

COO trades at a modest discount to Alcon and roughly at the peer median. It does not trade at a distressed multiple relative to its comp set. It trades at a large discount to its own history — COO carried a 22–30x forward multiple for most of 2015–2022 — but "cheap versus its own past multiple" is precisely the argument references/trade-construction.md warns is not evidence of anything, in either direction, and it is not used here as one.

Sum-of-the-parts — the CooperSurgical separation case, sized

Alcon — the nearest listed pure vision-care comparable — trades at 14.2x. So the market is currently valuing CooperVision at roughly a 6–11% EV/EBITDA discount to Alcon while ascribing a mid-teens multiple to CooperSurgical. That is not an obviously dislocated sum of the parts.

A separation is worth something, but the arithmetic is modest, not transformational. If CSI sells for $4.2bn gross / ~$3.8bn net, all $2.46bn of debt is retired and $1.4bn is repurchased at ~$75 (18.7m shares, 9.6% of the count), RemainCo is CooperVision on ~176m shares with no net debt. At Alcon's 14.2x on ~$960m of EBITDA that is ~$77/share; at 16.0x, ~$87/share. That range — +7% to +21% — is the honest size of the separation prize, and it depends almost entirely on the exit multiple.

---

4. Street bridge — required, not optional

Consensus estimates (Alpha Vantage EARNINGS_ESTIMATES, pulled 2026-07-28, cached to data/av_earnings_estimates_COO.json):

Revision trend. FY2026 EPS: $4.6294 now vs $4.6280 (30d) vs $4.6169 (90d) — flat to marginally up. FY2027 EPS: $5.0040 vs $4.9966 (30d) vs $5.0205 (90d) — down 0.3% over 90 days, up 0.1% over 30. Estimate revisions are flat. There is no PEAD-style revision momentum in either direction.

There is, however, one thing inside the revision data worth naming: near-term quarterly estimates have been cut while the full-year has not. FQ3-26 (Jul-31): $1.1203 now vs $1.1904 90 days ago, −5.9%. FQ4-26 (Oct-31): $1.1976 vs $1.2320, −2.8%. H1 beat and H2 was marked down — consistent with the company's own organic-growth guidance cut from 4.5–5.5% to 3.5–4.5% on 4 June.

Consensus rating and price target (public aggregators, July 2026): Moderate Buy, roughly 10 Buy / 5 Hold / 1 Sell. Average 12-month target $80.57 across 14 analysts, high $92, low $66 — an implied +11.8% to spot. Recent actions: Citigroup maintained Neutral and raised its target $69 → $76 on 8 July 2026; Goldman Sachs at $61; Needham $101; Barclays $103. The $61–$103 dispersion is unusually wide for a 15-analyst medtech name and is itself informative: the Street is not disagreeing about next year's numbers, it is disagreeing about the multiple and about the CooperSurgical outcome.

The required decomposition: is the gap numbers or multiple?

The house 12-month probability-weighted target is $77.45 against the Street's $80.57 — a 3.9% gap. It decomposes almost entirely to the multiple, not the numbers:

• Numbers: house FY2027E non-GAAP EPS $4.96 vs Street $5.004 — a 0.9% gap, which at a constant multiple is worth $0.71 of the $3.12 difference.
• Multiple: the residual $2.41 (77% of the gap) is the probability-weighted exit multiple. The house Base case uses 16.0x FY2027E EPS versus 15.5x today; the Street's $80.57 implies roughly 16.1x on its own $5.004 — essentially the same multiple. The real difference is that the house carries a 30% weight on a 12.0x bear outcome, which the Street's point-estimate targets do not express at all.

So the honest statement is: the house does not disagree with the Street about Cooper's earnings. It disagrees about the width of the distribution. That is a legitimate risk-management view. It is not a variant view, and Section 6 records that as a Criteria failure rather than dressing it up.

Risk if the Street is right. If CooperSurgical transacts at a genuine premium and CooperVision re-rates toward Alcon, the bear leg simply never happens, and the correct 12-month price is $90–100. Passing on this name costs real money in that world. That risk is named plainly and is the substance of calibration item B2/B7 below.

---

5. Factor & Anomaly Scorecard

All inputs computed from Alpaca SIP daily bars and SEC XBRL company facts. Single-name factor readings are noisy; these are evidence weights, not verdicts, and the read is stated for a contemplated LONG.

Synthesis. The scorecard is genuinely split, and it splits along an informative seam.

The balance-sheet and cash-quality factors are strong and unambiguous: accruals at −3.4% are the cleanest in four years, asset growth at +0.6% is near the bottom of the distribution, and gross profitability has risen every year for four years. These corroborate a capital-discipline and cash-generation mechanism.

The income-statement trajectory factors do not corroborate it: ROA is falling, gross margin is falling, the F-score is a middling 6/9, and momentum and revisions are both flat-to-negative.

Read together, they say something quite precise: this is a company whose cash and capital allocation are improving while its returns and margins on the reported P&L are not yet. That is exactly the profile of a business in the middle of a cost restructuring — and it is a supportive but not a confirmatory scorecard for a long. Per the Quality Criteria standard, the mechanism claim must be judged against this, and the honest verdict is that the quantitative evidence supports the cash mechanism and is silent-to-negative on the growth mechanism.

Calibration note on the momentum reading (item B1). CALIBRATION_WATCH B1 states that Momentum Criteria has never been tested on a name actually trading below its 200-day moving average, and that the GH/TXG/TWST experiment failed because all three turned out to be top-decile momentum names. COO is trading 1.3% below its 200-day MA with 12-1 momentum of −4.8%. It is a genuine, if mild, instance of the case B1 asks about — and Momentum Criteria does not bind here (Section 6). That is a data point on B1: at a 1.3% drawdown below trend the Criteria passes as neutral. It only bit on ISRG at −26%. B1 still needs a name deep below trend; COO narrows the question rather than answering it.

---

6. Valuation conclusion, kept separate from the other three

Per references/trade-construction.md, the four conclusions are stated separately and are not allowed to blend into one narrative.

1. Fundamental conclusion. Revenue growth is decelerating structurally — organic guidance has been cut from 6–8% to 3.5–4.5% across six consecutive quarters — while profitability and cash generation are improving materially through a cost reorganisation, a capex peak passing, and an M&A stop. Free cash flow roughly doubles from $434m (FY2025A) to a guided $600–625m (FY2026E) and to an implied ~$790m/yr in FY2027–28. Net debt/EBITDA falls from 2.3x to ~1.3x by FY2027E without any asset sale. The myopia-control growth engine is losing category share to spectacle lenses.

2. Expectations conclusion. The market expects essentially what the house model produces. ΔE = −0.4% on FY2027E revenue and −0.9% on FY2027E non-GAAP EPS. Consensus already embeds MiSight at its current 24% growth rate. The CooperSurgical strategic review is publicly announced, was the subject of four separate analyst questions on the last call, and sits comfortably inside a $61–$103 target dispersion. There is no material expectations gap in either direction.

3. Valuation conclusion. (v1.4.2 text, retained; see §0 for the authoritative two-horizon output.) Under the house forecast, COO is worth roughly $77–88 — DCF $87.74, scenario-weighted 12-month target $77.45, football-field centre of gravity high-$70s. Against $72.06 that is a +7% to +22% gap depending on method. Real, but not large, and the DCF portion of it is a statement about a 7.46% discount rate on a 76.7%-terminal-value stream.

Note the direction of the correction. The old five-year DCF marked COO up to $87.74, +21.8% above spot — the opposite of the systematic mark-down (16 of 16 targets below spot, item B16) the new method exists to fix. The new 12-month base is $77 (+7.1%), materially less dramatic in the other direction, and it contains no multiple opinion at all.

4. Portfolio conclusion — removed; this is not the memo's question.

SUPERSEDED (2026-07-29). The original read: "a modest, multiple-dependent valuation gap with no expectations gap is the textbook definition of a name that belongs on a watchlist rather than in a book." v1.6.0 outputs an analysis, not a position — whether this belongs in a book is a question about a particular book, and two books answer it differently.

What survives is the description, which is accurate and useful on its own: a modest, multiple-dependent valuation gap with no identifiable expectations gap. Under the new method the same fact set reads: the implied path requires 6.1% against 7.3% demonstrated (+1.3pp, the only PASS in this batch) — or −2.6pp and a FAIL once COO's own 16.7% operating margin replaces the scan's universe-wide 20.0% assumption (§0.1). The name sits on the boundary, and that is the finding.